Skip to content
Replication package for an eye-tracking study on code indentation
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Data and Analysis Scripts
Snippets
README.md

README.md

Data repository for results of paper "Indentations - A Simple Matter of Style or Program Comprehension?"

This repository contains the data we collected during our experiment examining the effect of indentation on program comprehension and visual effort.

Paper Abstract

An early study showed that indentation is not a matter of style, but provides actual support for program comprehension. In this paper, we present a non-exact replication of this study. Our aim is to provide empirical evidence for the suggested level of indentation made by many style guides. Following Miara and others, we also included the perceived difficulty, and we extended the original design to gain additional insights into the influence of indentation on visual effort by employing an eye-tracker. In the course of our study, we asked 22~participants to calculate the output of Java code snippets with different levels of indentation, while we recorded their gaze behavior. We could not find any indication that the indentation levels affect program comprehension or visual effort, so we could not replicate the findings of Miara and others. Nevertheless, our modernization of the original experiment design are a promising starting point for future studies in this field.

Descprition of Data

The data is splitted into the set of 22 participants that saw a randomized order of indentation and program snippet and the set which additionally contains those 17 participants that saw the same order. For our paper, we only included the set 22 'random' trials into our analysis.

For each set there are two folders TrialData and Analysis:

Trial Data

The raw data contains for each participant with his/her ID:

  • The eye tracking data in the file Gaze_ID
  • General information such as age, gender, etc. about the participant and the trial in general in the file GeneralInfo_ID
  • The positions of the snippets in the rating with equal and real indentations in RatingsEqualIndentations_ID and RatingsRealIndentations_ID respectively
  • The result and time for each snippet in Results_ID

Analysis data

These files contain the summarized information of all participants of the respective set:

  • For the gaze analysis of our data, we imported the gaze data of the participants into Ogama and calculated the fixations and saccades with this tool. The result of this analysis is stored in CalculatedGaze
  • The general info of all participants is stored in "GeneralInfo"
  • The ratings with equal and real indentations are stored in RatingsEqualIndentations and RatingsRealIndentation respectively
  • The results of all participants are stored in Results

Analysis

The following table shows the results of the statistical analysis:

Attribute Test 22 (random) 39 (all)
FixationDurationMean(log-transformed) ANOVA F-value 3,06
p-value 0,03
eta-square 0,04
Shapiro-Wilk W-value 0,35
p-value 0,00
Shapiro-Wilk (untransformed) W-value 0,99 0,79
p-value 0,57 0,00
Friedman Chi-square 15,93
p-value 0,00
FixationDurationMedian ANOVA F-value 2,85 1,64
p-value 0,04 0,18
eta-square 0,03
Shapiro-Wilk W-value 0,61
p-value 0,00
Shapiro-Wilk (original) W-value 0,98 0,50
p-value 0,18 0,00
Friedman Chi-square 16,04
p-value 0,00
Time (log-transformed) ANOVA F-value 0,44 2,24
p-value 0,72 0,09
Shapiro-Wilk W-value 0,98 0,99
p-value 0,10 0,16
Shapiro-Wilk (untransformed) W-value 0,87
p-value 0,00
SaccadicAmplitude(log-transformed) ANOVA F-value 1,67
p-value 0,18
Shapiro-Wilk W-value 0,98 0,87
p-value 0,11 0,00
Shapiro-Wilk (untransformed) W-value 0,95 0,50
p-value 0,001 0,00
Friedman Chi-square 13,61
p-value 0,00
Correctness Friedman Chi-square 3,23 0,36
p-value 0,36 0,95
RatingReal Friedman Chi-square 5,35 18,51
p-value 0,15 0,00
FixationRate Friedman Chi-square 7,36 14,88
p-value 0,06 0,00
Shapiro-Wilk W-value 0,76 0,27247
p-value 0,00 0
Shapiro-Wilk (log) W-value 0,56 0,71
p-value 0,00 0,00

The analysis was performed with R. See the *Eval.R" script for some examples.

You can’t perform that action at this time.